|
|
https://www.ispreview.co.uk/index.php/2018/04/peak-t...
A new study into the impact on broadband ISP download speeds of peak (afternoon) vs off-peak (morning/daytime) internet usage at the largest six broadband providers in the United Kingdom has revealed that Plusnet suffers the biggest proportional drop in performance, while oddly TalkTalk got faster.
jelv
AAISP November 2016
(Previous ISP Plusnet November 2001 to October 2016) Why I left Plusnet
Telephone rental: Pulse8
|
|
|
|
|
|
The author of the above post is a thinkbroadband staff member. It may not constitute an official statement on behalf of thinkbroadband.
|
|
|
Seems to disagree with TBB analysis of the data that ISPReview have used (ie TBBs data in the first place). TBB article here.
Don't have time to analyse why it is so different but from a quick view the TBB results are almost the opposite of the ISPReview conclusion.
|
|
Register (or login) on our website and you will not see this ad.
|
|
|
The people testing during the day are not the same people testing at night...
Different mixes of technology and line lengths.
The presentation I use is designed to mitigate that effect.
|
|
The author of the above post is a thinkbroadband staff member. It may not constitute an official statement on behalf of thinkbroadband.
|
|
|
Hi Andrew,
So am I right in thinking 10 customers testing during the day, all get high speeds as they have 80Mbit connections.... 10 different customers test in the evening and get 20Mbit as they�re on longer lines....
This would be reported a huge difference in peak congestion even though all connections are working to their maximum?
Kris
Sky Fibre
Ashington (Northumberland) Exchange
|
|
|
Mark has taken the httpx6 figures from two tables and presented them side by side Ian. They correspond.
I think it's a bit cheeky not to acknowledge his source, even if he asked permission in the first place. The first place I went after reading it from jelv's link was our Main Site.
My broadband basic info/help site - www.robertos.me.uk. Domains, site and mail hosting - Tsohost.
Connection - AAISP Home::1 80/20. 200GB. Sync 69311/13513Kbps @ 600m. BQMs - IPv4 & IPv6 |
|
|
Ten users, maybe. Thousands, nope. It shows a ranking, not an estimate for you or me.
My broadband basic info/help site - www.robertos.me.uk. Domains, site and mail hosting - Tsohost.
Connection - AAISP Home::1 80/20. 200GB. Sync 69311/13513Kbps @ 600m. BQMs - IPv4 & IPv6 |
|
|
On top of that, it misled me into thinking it was from a newer ofcom report, as he showed 2016 ofcom data, then used data from new monthly tbb data so two totally different sources.
|
|
|
The way I have presented the numbers that does not happen, the way ISPreview did things that is possible
|
|
The author of the above post is a thinkbroadband staff member. It may not constitute an official statement on behalf of thinkbroadband.
|
|
|
I�m still thinking we�re reading far to much into the data than we should....
Basically for each ISP we�re taking the average speed test during the day and the average for the evening, and the difference is been treated as peak time slow down.
Surely for this to be true, we would need the distribution of connection speeds who test during the day to match that run in the evening?
Kris
Sky Fibre
Ashington (Northumberland) Exchange
|
|
|
I�ve read your article again now and can now see that this wasn�t what you suggested at all!
Because I read the ISP review first,I had it that the whole point of the thinkbroadband article was a comparison between peak and off peak, when in reality it�s 2 independant datasets.
I�ve a Samknows white box and I think they�re the only way you could compare peak and off peak (as each line would be tested at both times)
Kris
Sky Fibre
Ashington (Northumberland) Exchange
|
|
|
If you use the ISPreview interpretation which is WRONG
I suggest you READ https://www.thinkbroadband.com/news/8019-peak-and-of...
I am definitely not saying TalkTalk is 14% faster at night.
The purpose of my article is to show the differences that arise from single thread (very sensitive to congestion and other issues) versus multiple thread.
|
|
The author of the above post is a thinkbroadband staff member. It may not constitute an official statement on behalf of thinkbroadband.
|
|
|
The article does say: Nevertheless we have done a bit of new analysis using Thinkbroadband�s latest Q1 2018 (March) consumer speedtest data for BT, Plusnet, EE, Sky Broadband, TalkTalk and Virgin Media to check. But that doesn't look like proper attribution to me but that may be acceptable to MrS and I'm sure if it isn't then he will be in touch with them.
|
|
|
I didn�t notice that.
My broadband basic info/help site - www.robertos.me.uk. Domains, site and mail hosting - Tsohost.
Connection - AAISP Home::1 80/20. 200GB. Sync 69311/13513Kbps @ 600m. BQMs - IPv4 & IPv6
|
|
|
I get a slight slowdown (< 10%) early on some weekday evenings but that is all, it's not enough to impact my use at all.
Kevin
plusnet Unlimited Fibre Extra - sync 75433/20000 at around 450m - BQM
Using OpenDNS
Domains and web hosting with TSOHOST |
|
|
|
I had to carefully scan the article to see if they had attributed it - it isn't really obvious and I would expect much better when using someone else's data.
|
|
|
|
If anything I've noticed a slowdown when accessing some larger websites, rather than a general slowdown night or day.
|
Pages in this thread: 
Print Thread
jelv
